nf12 open-sources lossless 12-bit compression of BF16 models, runs on Ampere through Blackwell
AlpinDale · x · 2026-11-07
dphnAI open-sourced nf12, a lossless fixed-length 12-bit encoding of BF16 weight matrices based on Together AI's NyoomFloat12 paper (Liberman, Fang, Zhang, Dao, Fu, 2026). Since no hardware natively supports this dtype, the author focused on runtime efficiency; the bit/memory layout (tiled) differs from the paper's spec. It targets NVIDIA GPUs from Ampere up to Blackwell, requiring SM80+, CUDA 13.0+, and Python 3.10+.
Related event: Developer open-sources nf12: lossless 12-bit compression for BF16 models(10 posts)→
More from Infra
- US AI scale now hinges on coordinating power, components, agentic workflows and humanoid pilots — ingliguori · 2026-11-07
- Fudan's D3D diffusion drafter enables async speculative decoding, up to 7.48x speedup on Qwen3 — pmttyji · 2026-11-07
- Study: data centers have only marginal impact on residential electricity prices — TinfoilTricorn · 2026-11-07
- YouJiacheng: post-training needs high tps for long-tail rollouts — YouJiacheng · 2026-11-07
- vLLM contributor: lossless KV cache quantization viable but not worth it — AlpinDale · 2026-11-07
- NVIDIA's first co-packaged optics Ethernet switch, Spectrum-X Photonics, enters full production — TinfoilTricorn · 2026-11-07